By Dr. Rob Urban
TL;DR:
- The Crux: Businesses routinely mistake surface-level symptoms for root causes, wasting time and money fixing the wrong problems based on assumptions rather than evidence. Analytics acts as a neutral investigator that replaces guesswork with hard proof to reveal hidden friction points.
- Descriptive Analytics (What Happened?): Establishes baseline reality and forces everyone to agree on facts rather than opinions (e.g., recognizing traffic didn’t drop, but emails were landing in spam).
- Diagnostic Analytics (Why Did It Happen?): Pinpoints the underlying root cause behind a symptom (e.g., discovering cart abandonment was caused by surprise shipping fees at checkout rather than high product prices).
- Predictive Analytics (What Will Happen Next?): Uses historical pattern recognition to anticipate upcoming shifts rather than being caught off guard by recurring cycles (e.g., spotting early indicators of seasonal demand before the rush arrives).
- Prescriptive Analytics (What Should We Do?): Translates data insights into decisive, meaningful actions that remove operational excuses and eliminate wasted effort.
Estimated Read Time: 12–14 minutes
Treating Symptoms vs. Diagnosing Problems
One of the things I have noticed over the years is that businesses almost always describe the symptom before they describe the problem. They will tell me leads are down, sales are slowing, customers are not calling, or the website just does not seem to be working anymore. Those are all perfectly valid observations, but they are no more useful than walking into your doctor’s office and announcing that your knee hurts. It is a starting point, not a diagnosis.
A sore knee might mean you twisted it playing pickleball. It might mean arthritis. It might mean you finally accepted your age by attempting to move a refrigerator without asking for help. The symptom is real, but the cause remains unknown. Analytics exists because businesses have a habit of treating symptoms as though they are explanations, and those two things are vastly different.
I think that is one of the reasons so many companies spend money in the wrong places. They assume the first thing they notice must also be the thing that needs fixing. If sales fall, they buy more advertising. If website traffic drops, they redesign the homepage. If customers complain about pricing, they lower prices. Every one of those decisions might eventually prove to be the right one, but making them before understanding the underlying problem is a little like replacing your roof because you found water on the kitchen floor.
Maybe the roof really is leaking. Or maybe your twelve-year-old just discovered that ice cubes slide surprisingly well across hardwood floors. Until you know where the water came from, you are just spending money in increasingly creative ways.
One of my favorite consulting engagements started with an owner who was absolutely convinced Google had done something to his business. He did not know exactly what, but he knew Google was involved because traffic had fallen and the phones were not ringing. To be fair, Google changes its algorithm often enough that blaming Google is not completely irrational. It has become the digital version of blaming the weather. Sometimes it is actually responsible. Most of the time it is simply convenient because nobody can call Google and ask to speak to the manager.
We started looking through the analytics together, expecting to find some dramatic collapse in rankings or a major technical issue. Instead, something much more interesting appeared. Organic traffic had barely changed. Rankings were stable, and in some cases, they had actually improved. People were still finding the company, reading the service pages, and spending several minutes on the site. The phones were not ringing because people were clicking the email address instead, and every single one of those emails had been landing in a spam folder for nearly six weeks. The owner had been preparing to spend tens of thousands of dollars rebuilding a website that was not broken because a perfectly good email inbox had quietly decided every new customer looked suspicious.
Moments like that are why I enjoy analytics so much. They remind me that businesses are often standing three feet away from the answer while searching everywhere else. It is human nature. We naturally assume complicated problems require complicated solutions. We like dramatic stories, believing our competitors have discovered some secret strategy or that a mysterious algorithm has singled us out for punishment.
The truth is usually much less exciting. Most business problems are not caused by grand conspiracies. They are caused by tiny little friction points that nobody noticed because everyone was too busy searching for something bigger.
The Misleading Nature of Human Intuition
Think about the last time you lost your car keys. I doubt your first thought was to carefully retrace your steps. Most of us immediately begin inventing increasingly ridiculous explanations. Maybe you left them at the grocery store. Maybe they fell out in the parking lot. Maybe somebody accidentally picked them up. Maybe the dog somehow carried them into the backyard. Twenty minutes later, you discover they are sitting exactly where you put them, which is both reassuring and mildly insulting because apparently the greatest obstacle to finding your keys was your own confidence.
Businesses do this every day. A contractor assumes homeowners suddenly stopped spending money when the estimate request form simply stopped working on mobile devices. A law firm spends months wondering why consultation requests have slowed without realizing the page takes twelve seconds to load because someone uploaded photographs large enough to be visible from space. A retailer blames declining foot traffic on the economy even though customers have been complaining for six months that nobody ever answers the phone.
None of those owners are unintelligent. Quite the opposite. They are experienced, successful people who simply did what all of us do from time to time: they reached for the explanation that felt right instead of the one the evidence supported.
That is where analytics quietly changes the conversation. Instead of asking what we think happened, it asks what we can actually prove. That sounds like a small difference, but it completely changes the quality of the decisions that follow.
Opinions are still welcome. Experience still matters. Intuition still has tremendous value because it often points you in the right direction. The difference is that intuition gets to make the opening argument instead of the final decision, leaving analytics to act as the jury.
That is an important distinction because people sometimes talk about analytics as though it is replacing human judgment, and I do not think that is true at all. In my experience, analytics actually makes experienced business owners more valuable because it gives them something they rarely had decades ago: immediate feedback. A great contractor still knows construction. A great physician still knows medicine. A great attorney still understands the law. Analytics does not replace expertise any more than an X-ray replaces a doctor. It simply gives smart people better information to work with, and better information almost always leads to better decisions.
The businesses that consistently outperform their competitors are not necessarily the ones with the biggest advertising budgets or the fanciest software. More often than not, they are the businesses that stay curious a little longer than everyone else. They resist the temptation to grab the first explanation that sounds reasonable. They keep asking questions after everyone else thinks they have found the answer. They understand something detectives, doctors, and good mechanics have known forever: the first explanation is sometimes correct, but the second or third one is often where the truth has been hiding all along.
Understanding the Diagnostic Progression
One of the things that makes analytics seem intimidating is that people insist on giving perfectly ordinary ideas very impressive names. Somewhere along the way, someone decided it sounded much smarter to say “descriptive analytics” than “figuring out what happened,” and ever since then we have all been pretending those are different things. They are not.
Think about any mystery you have ever tried to solve. It does not have to be business-related. Maybe you walked into the kitchen and discovered your dog looking strangely innocent while an entire rotisserie chicken had somehow vanished from the counter. Nobody begins that investigation by asking what the dog is likely to steal next Tuesday. You start with the obvious question: What happened?
Once you have established that the chicken has indeed disappeared, you move to the next question: How did it happen? Then you start wondering whether this furry criminal has been running the same operation every Thanksgiving. Finally, you decide what changes need to be made, which usually involve moving future chickens somewhere well above nose level. Without realizing it, you have just worked through the exact framework businesses use to make sound decisions.
Business works the same way, though hopefully with fewer stolen chickens.
Every meaningful decision begins with understanding what actually happened. That sounds almost insultingly simple until you realize how often companies skip that step entirely. I cannot tell you how many times I have heard someone say their website is not working, when what they really mean is they have not looked closely enough to know what the website is actually doing. That is a little like walking into your garage, seeing your car will not start, and announcing that the engine is ruined before you have checked whether there is any gas in the tank. It might be the engine, or it might be that your teenager returned the car with just enough fuel to coast into the driveway.
The first job of analytics is remarkably unglamorous. It simply forces everyone to agree on reality before anyone starts debating solutions. You would think that would be automatic, but spend enough time around businesses and you will discover that reality is surprisingly negotiable.
Ask five people why sales dropped, and you will often receive six different answers. Ask them how many qualified leads actually came through the website last month, and suddenly everyone starts looking at each other like students caught off guard by a pop quiz.
Numbers do not settle every argument, but they eliminate an awful lot of imaginative storytelling.
I worked with a retailer years ago whose owner was convinced customers had become more price-sensitive. He had pages of handwritten notes about the economy, inflation, competitors, and changing buying habits. He had highlighted newspaper articles, detailed charts, and enough supporting material to defend a doctoral dissertation on consumer confidence. The only thing he did not have was evidence that any of those theories applied to his own business.
When we looked at the analytics, we discovered something almost embarrassingly simple. Customers were not abandoning their shopping carts when they saw the price tag. They were abandoning them when shipping costs appeared on the final checkout screen. They had no objection to buying the product; they simply objected to feeling surprised. The issue was not pricing, it was timing. One small change to how shipping was presented made a measurable difference almost immediately.
That is one of my favorite things about analytics. Every once in a while, it reminds you that customers are wonderfully logical, even when businesses assume they are being unpredictable. People do not mind paying for value, but they mind unpleasant surprises. They do not necessarily leave because something costs too much; they leave because they no longer trust what comes next. Analytics cannot read minds, but it can point toward the exact moments where trust quietly begins to leak out of the customer experience.
Moving Beyond Surface-Level Metrics
Knowing what happened is useful, but understanding why it happened is where the real value begins to appear. Imagine walking into your doctor’s office after your annual physical. He looks at your blood pressure, nods thoughtfully, and says it is higher than last year before wishing you a nice day. Technically, he answered the first question by noting that something changed. Unfortunately, that is also the least helpful part of the conversation. You did not schedule the appointment to discover that a number moved. You scheduled it because you wanted someone to explain why it moved and what should be done about it.
Businesses deserve the same level of curiosity. If website traffic falls, do not stop at the graph. Ask what changed. Did rankings decline? Did search behavior shift? Did a competitor publish better content? Did someone accidentally block search engines from indexing an important page? Did your developer proudly clean things up on Friday afternoon before leaving for vacation? That last scenario happens often enough that I no longer consider it hypothetical.
One memorable example involved a company that watched website traffic decline over several months and became convinced their industry was simply slowing down. It was a comforting explanation because it suggested there was nothing anyone could have done differently. Markets change, economies fluctuate, and sometimes that is just life.
The problem was that the analytics kept telling a different story. Organic traffic was declining, but branded searches, direct visits, and referrals from existing customers remained almost identical. The audience had not disappeared; new customers simply were not discovering the company anymore.
A closer look revealed that dozens of older blog articles had quietly lost rankings after years of neglect. Nobody had updated or improved them while competitors published newer, more useful resources that answered today’s questions instead of yesterday’s. The business did not have a market problem. It had a content problem masquerading as an economic problem.
Numbers are just footprints. The interesting part is figuring out who left them, where they were going, and why they suddenly changed direction. The spreadsheet is not the story. It is merely the trail of breadcrumbs leading toward it.
Behavior over Opinion
Good business owners and good detectives have far more in common than either group would probably admit. Neither one is paid to admire the evidence; they are paid to make sense of it. A detective does not stand over a footprint and congratulate himself for finding it. He wants to know who left it, where they came from, where they went, and whether they are likely to return. Every number is simply a clue that points toward a much more interesting question.
I remember talking with the owner of a service company who proudly announced that his website traffic had doubled over the previous year. The graphs looked fantastic, with every line climbing in the right direction. Then I asked the question that has a funny way of ruining celebrations: How many more customers did you get?
He looked at me for a moment, glanced back at the screen, and admitted he was not sure. We have become so accustomed to celebrating the numbers that are easiest to measure that we sometimes forget to measure the ones that actually keep the lights on.
Traffic by itself has about as much value as counting how many people walk through the front door of a grocery store without bothering to notice whether anyone bought milk. Imagine if a restaurant owner proudly announced three thousand people came through the doors this week, only to quietly admit that only forty of them ordered dinner. You would not congratulate him on his foot traffic. You would start wondering why everyone else left hungry.
The same thing happens online every single day. Businesses become obsessed with attracting visitors while paying remarkably little attention to what happens after they arrive.
Websites are not just collections of pages; they are conversations. Every page is answering a question, every button is making a promise, and every element is either building trust or quietly eroding it. Customers do not wake up hoping to browse service websites for fun. They have a problem, and your website is joining a conversation that started long before they found you.
This is where analytics feels less like software and more like psychology. Human beings are predictable in specific ways. We want answers quickly, we become suspicious when something feels intentionally vague, and we get nervous when a website asks for contact information before earning the right to have it.
None of those reactions show up in customer surveys because most people never fill them out. They show up in behavior: abandoned forms, repeated searches, long pauses on certain pages, sudden exits, and unexpected spikes in traffic after you answer a question everyone has been asking.
If someone tells me they loved a website, but the analytics show they stayed for twelve seconds before leaving, I am inclined to believe the stopwatch.
The Pitfall of Preemptive Solutions
Businesses tend to fall in love with solutions long before they have earned them. We love buying software, redesigning websites, creating new logos, launching advertising campaigns, and viewing colorful dashboards. None of those things are inherently bad, but the problem begins when the solution arrives before the diagnosis. That is how companies end up spending fifty thousand dollars solving a five-hundred-dollar problem.
Analytics has a wonderful habit of separating genuine emergencies from expensive overreactions. It will not prevent every mistake, but it dramatically reduces the number of decisions made simply because someone had a bad feeling after a Tuesday sales meeting.
As technology evolves, the quality of the answer will always depend on the quality of the question. Artificial intelligence and machine learning can process massive amounts of information, but they cannot rescue a business that is asking the wrong question. If you spend six months trying to figure out how to attract more visitors while completely ignoring why your existing visitors never become customers, you have simply found a faster way to head in the wrong direction.
The companies that get the most out of data are the ones willing to be proven wrong. They understand that analytics is not there to confirm their opinions, but to challenge them. Once you stop treating data as a report card and start treating it like a conversation with your customers, you become less defensive, more curious, and far more likely to discover opportunities your competitors are still arguing about in conference rooms.
Curiosity keeps experience, confidence, and instinct from becoming liabilities. The moment you believe you have seen everything, customer habits shift, technology moves, and buying behaviors evolve. Analytics gives you a chance to notice those changes before they become expensive mistakes.
Predicting Trends and Taking Action
Predictive analytics is not magic; it is pattern recognition. If you live in a region where afternoon summer storms are guaranteed, you do not need to be a meteorologist to know when to put the top up on a convertible. You are simply comparing current conditions with past patterns and making an educated prediction.
Patterns allow businesses to stop reacting to yesterday and start preparing for tomorrow.
Consider a seasonal business that experiences an overwhelming rush every spring, panics, hires late, and struggles through delays, only to slow down every autumn. When you examine years of historical data, the sequence is obvious: phone calls spike first, website traffic follows, estimate requests climb two weeks later, and contracts peak shortly after. The business receives an early warning signal every year, but fails to act because everyone is too busy putting out immediate fires to notice the pattern. Once that pattern is understood, hiring happens earlier, marketing aligns with demand, and operational stress drops significantly.
The ultimate goal of all this data is prescription: deciding what to do next. Analytics becomes useless when it ends in massive reports that yield no real organizational changes. Good analytics should occasionally make you uncomfortable because it forces decisions. If data reveals that eighty percent of your revenue comes from twenty percent of your services, or that customers consistently drop off at a specific step, a decision must be made.
The hardest part of data analysis is rarely interpreting the numbers. It is convincing people to let go of legacy ideas, products, or processes they have grown emotionally attached to over time.
Analytics acts like a brutally honest friend. It does not care how much you love a specific page on your website if visitors ignore it. It does not care about traditional advertising channels if they no longer produce leads. It simply reports the objective reality of customer behavior.
When it comes to artificial intelligence, the technology excels at processing data and identifying complex patterns, but it lacks the contextual understanding that comes from human experience. AI can spot a pattern, but experienced leaders understand why that pattern matters. The businesses gaining the largest return from modern technology are not using it to replace human judgment, but to amplify it. Technology makes your operations faster, but speed only helps if you are already driving in the right direction.
At the end of the day, analytics is not about complex algorithms, massive spreadsheets, or chasing the newest software trend. It is about having the discipline to separate what is actually happening from what you assume is happening.
If you want to make better decisions, stop trying to fix the first symptom that catches your attention. Ask better questions, follow the evidence wherever it leads, and build your business strategy on proven reality rather than comfortable guesses.
Robert Urban